The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill recommends installing the 'uv' package manager using a piped shell command (curl -LsSf https://astral.sh/uv/install.sh | sh). This targets a well-known technology provider and follows the official installation method for that tool.
[COMMAND_EXECUTION]: The script scripts/check_env.py executes system commands like git, ffmpeg, and uv to verify the local environment. These commands are hardcoded in the script and do not incorporate untrusted user input.
[DATA_EXPOSURE]: The skill includes logic to read the local .env file to identify unconfigured placeholders (e.g., 'YOUR_KEY_HERE'). This is part of a diagnostic workflow. The skill also provides extensive documentation and scripts (pii_guard.md, sanitize_history.sh) to prevent the accidental exposure or exfiltration of credentials and PII.
[DYNAMIC_EXECUTION]: The skill implements a SessionStart hook mechanism in .claude/settings.json that executes a local bash script (session-start-check.sh) when the agent enters the repository. The script is used for non-malicious purposes, specifically to provide environment-related notifications to the user.

auto-repo-setup